The central serotonergic (5HT) system is one of the candidate systems that may mediate the abnormal mood and behavioral response to ovarian steroids observed in women with menstrual-related mood disorders (MRMD). Numerous studies have attempted to measure 5HT system function in women with MRMD and controls; however, previous studies were restricted to the evaluation of indirect measures of 5HT function (e.g., platelet 5HT uptake or responses to serotonergic pharmacologic challenges). Radioligand imaging of specific components of the 5HT system currently represents the most direct measure of central serotonergic function available. In this study we will image serotonergic measures during pharmacological conditions previously demonstrated to induce a remission and a subsequent provocation of mood symptoms in women with MRMD but not controls.

This protocol has two outcome measures (5HT(1A) binding and SERT binding), two groups of subjects (women with MRMD and controls), and three distinct hormonal conditions (hypogonadism, estrogen replacement, and progesterone replacement). Our principal aim in this study is to investigate differences in the outcome measures in women with MRMD compared to controls. Our primary hypothesis is that 5HT(1A) binding will be decreased and SERT binding will be increased in women with MRMD. A secondary hypothesis is that women with MRMD, but not controls, will display reproductive steroid modulation of 5HT(1A) and SERT binding. This study will help test existing hypotheses about the pathophysiologic relevance of serotonin function in MRMD and the role of reproductive steroids in serotonergic regulation. First, we selected 5HT(1A) receptor binding as an outcome measure for the following reasons: 5HT(1A) function is modulated by both estradiol and progesterone, abnormalities of 5HT(1A) binding have been reported in both depressive and anxiety disorders as well as possibly MRMD, 5HT(1A) receptors have been identified as potential mediators of estradiol's antidepressant-like effects in the forced swim test, and the 5HT(1A) system is involved in the regulation of GABA activity abnormalities of which have been implicated in MRMD. Second, we selected SERT binding as an outcome measure for the following reasons: selective serotonin reuptake inhibitors, but not traditional antidepressants, are effective treatments for MRMD; abnormalities of SERT binding have been reported in depressive disorders in women, preliminary data suggest that women with MRMD are characterized by an elevated frequency of the long, more transcriptionally active polymorphism of the serotonin transporter (5HTTLPR) (Roca et al); women with MRMD have altered imipramine binding and platelet 5HT uptake compared to controls as well as altered platelet paroxetine binding (which normalize with successful treatment with GnRH agonist); and, finally, SERT message, binding and protein have been reported to be changed by ovarian steroids in preclinical studies.

We hypothesize that the effects of hormonal condition on 5HT(1A) (receptor) binding will differ in patients and controls. Specifically, we anticipate that 5HT1A binding will be decreased in hippocampus and prefrontal cortex in patients with MRMD but not controls. Such a decrease, if a trait characteristic, will be demonstrable irrespective of hormonal state. However, as symptoms of MRMD are directly precipitated by estradiol or progesterone (see below), it is possible that symptoms are accompanied by state and steroid-dependent changes in 5HT(1A) binding and function. Consequently, our secondary hypothesis is that estradiol or progesterone induces changes in 5HT(1A) (receptor) binding in women with MRMD but not in control women.

Similarly, we hypothesize that women with MRMD will have increased SERT binding activity compared to controls, consistent with the therapeutic efficacy of selective serotonin reuptake inhibitors in this condition. However, a secondary hypothesis is that estradiol or progesterone-induced changes in SERT binding will occur in a hormone-dependent and mood state-dependent manner. Specifically, consistent with the literature in rodents, acute changes in ovarian steroids will increase SERT binding in association with the induction of symptoms in women with MRMD but not controls.